REVIEW 17 cited by
FlexGen: High-Throughput Generative Inference of Large Language Models with a Single GPU
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
FlexGen: High-Throughput Generative Inference of Large Language Models with a Single GPU
read the original abstract
The high computational and memory requirements of large language model (LLM) inference make it feasible only with multiple high-end accelerators. Motivated by the emerging demand for latency-insensitive tasks with batched processing, this paper initiates the study of high-throughput LLM inference using limited resources, such as a single commodity GPU. We present FlexGen, a high-throughput generation engine for running LLMs with limited GPU memory. FlexGen can be flexibly configured under various hardware resource constraints by aggregating memory and computation from the GPU, CPU, and disk. By solving a linear programming problem, it searches for efficient patterns to store and access tensors. FlexGen further compresses the weights and the attention cache to 4 bits with negligible accuracy loss. These techniques enable FlexGen to have a larger space of batch size choices and thus significantly increase maximum throughput. As a result, when running OPT-175B on a single 16GB GPU, FlexGen achieves significantly higher throughput compared to state-of-the-art offloading systems, reaching a generation throughput of 1 token/s for the first time with an effective batch size of 144. On the HELM benchmark, FlexGen can benchmark a 30B model with a 16GB GPU on 7 representative sub-scenarios in 21 hours. The code is available at https://github.com/FMInference/FlexGen
Forward citations
Cited by 17 Pith papers
-
RoPE-Aware Bit Allocation for KV-Cache Quantization
Block-GTQ performs RoPE-aware greedy bit allocation on KV caches using per-block energy scores, cutting logit MAE 32-80% versus uniform TQ-MSE and lifting long-context task scores substantially at 2-3 bits per dimension.
-
SiDP: Memory-Efficient Data Parallelism for Offline LLM Inference
SiDP distributes model weights across a DP group with WaS and CaS modes to increase KV cache capacity by up to 1.8x and end-to-end throughput by up to 1.5x over vLLM on H20/H200/B200 GPUs for offline LLM inference.
-
Reference-Augmented Learning for Precise Tracking Policy of Tendon-Driven Continuum Robots
Reference-augmented learning with RNN surrogate and stochastic perturbations cuts average position error by 50.9% for 6-DOF tracking on a three-section TDCR compared to non-augmented baselines.
-
Efficient Memory Management for Large Language Model Serving with PagedAttention
PagedAttention achieves near-zero waste in LLM key-value cache memory and enables 2-4x higher serving throughput than prior systems.
-
What to Keep, What to Forget: A Rate--Distortion View of Memory Compaction in LLMs and Agents
KV-cache eviction, prompt compression, recurrent state bounding, and agent memory consolidation are unified as one rate-distortion problem with a shared lower bound, shared failure mode, and transferable mechanisms.
-
From Rigid to Dynamic: Entropy-Guided Adaptive Inference for Long-Context LLMs
EntropyInfer adaptively allocates inference compute using per-head attention entropy for rigid/dynamic classification during prefilling and compresses KV cache with generated tokens, achieving up to 2.39x speedup on l...
-
Do Transformers Need Three Projections? Systematic Study of QKV Variants
Q-K=V projection sharing in transformers matches standard QKV performance with 50% KV cache reduction and combines with GQA/MQA for up to 96.9% reduction across vision and language tasks.
-
Motion-Compensated Weight Compression
MCWC aligns permutation-symmetric blocks across layers to enable sequential prediction and residual entropy coding, improving rate-accuracy tradeoffs versus quantization and prior codecs on language and vision models.
-
NVLLM: A 3D NAND-Centric Architecture Enabling Edge on-Device LLM Inference
NVLLM offloads FFN computations to integrated 3D NAND flash with page-level access and keeps attention in DRAM, delivering 16.7x-37.9x speedups over GPU out-of-core baselines for models up to 30B parameters.
-
AsyncTLS: Efficient Generative LLM Inference with Asynchronous Two-level Sparse Attention
AsyncTLS delivers full-attention accuracy with 1.2-10x operator speedups and 1.3-4.7x end-to-end throughput gains on 48k-96k contexts via two-level sparse attention and asynchronous offloading.
-
H$_2$O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models
H2O evicts non-heavy-hitter tokens from the KV cache using a dynamic submodular policy, retaining recent and frequent-co-occurrence tokens to reduce memory while preserving accuracy.
-
AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration
AWQ quantizes LLM weights to low bits by scaling salient channels based on activation statistics, outperforming prior methods on language, coding, math, and multi-modal benchmarks.
-
GSRQ: Gain-Shape Residual Quantization for Sub-1-bit KV Cache
GSRQ applies a gain-shape variant of K-means inside residual quantization to improve directional fidelity, raising LongBench accuracy from 11.34 to 33.54 at 1-bit on LLaMA-3-8B.
-
DisagFusion: Asynchronous Pipeline Parallelism and Elastic Scheduling for Disaggregated Diffusion Serving
DisagFusion achieves 3.4x-20.5x higher throughput and 18.5x lower latency for diffusion serving via asynchronous pipeline parallelism and elastic hybrid scheduling on disaggregated hardware.
-
TIDE: Efficient and Lossless MoE Diffusion LLM Inference with I/O-aware Expert Offload
TIDE schedules I/O-aware expert offloading for MoE diffusion LLMs by solving for an optimal refresh interval that exploits temporal stability of activations, yielding up to 1.5x throughput gain losslessly.
-
Position: LLM Inference Should Be Evaluated as Energy-to-Token Production
LLM inference should be reframed and evaluated as energy-to-token production with a Token Production Function that accounts for power, cooling, and efficiency ceilings.
-
Reference-Augmented Learning for Precise Tracking Policy of Tendon-Driven Continuum Robots
A reference-augmented offline learning framework for 6-DOF tracking control of tendon-driven continuum robots achieves 50.9% lower average position error than non-augmented baselines.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.